A new algorithm has enabled the Himawari-8 geostationary meteorological satellite to retrieve ocean color information every 10 minutes, providing scientists with a more detailed view of rapidly changing coastal waters.

By learning from high-quality Moderate Resolution Imaging Spectroradiometer (MODIS) ocean color observations, the transformer-based model overcomes key limitations of Himawari-8, including its low signal-to-noise ratio, and reduces retrieval errors compared with standard hourly products.

The new approach provides a promising tool for near-real-time monitoring of fast-changing ocean processes, including algal blooms, sediment transport, and water quality variations across the Asia-Pacific region.

Ocean color remote sensing is essential for observing marine ecosystems, primary productivity, coastal water quality, and harmful algal blooms. Polar-orbiting satellites, such as MODIS, provide accurate ocean color observations, but they usually pass over the same region only once or twice a day. Consequently, short-term changes in dynamic coastal environments can often go undetected.

Geostationary satellites offer a complementary advantage. Positioned above the same region, Himawari-8 can observe the Earth at a very high temporal frequency. However, it was designed as a meteorological satellite rather than a dedicated ocean color sensor. Its relatively low signal-to-noise ratio and the use of standard hourly composites can introduce systematic biases, including underestimation of remote sensing reflectance (Rrs) in turbid waters and overestimation in clearer waters.

To address this challenge, researchers from the Aerospace Information Research Institute (AIR) of the Chinese Academy of Sciences (CAS), Inner Mongolia Normal University, the University of Oslo, and other partner institutions developed a transformer-based algorithm capable of retrieving Rrs from Himawari-8 multispectral data at a 10-minute resolution. The study was recently published in the Journal of Remote Sensing.

The algorithm combines a conventional atmospheric correction framework with a transformer neural network. After correcting for gas absorption and Rayleigh scattering, the model learns the nonlinear relationship between top-of-atmosphere reflectance and ocean surface reflectance. Its inputs include solar viewing geometry, six Himawari-8 reflectance bands, aerosol optical thickness, and wind speed. High-quality MODIS Aqua Rrs products and AERONET-OC in situ measurements were used as reference data for training and validation.

Validation results demonstrated that the new algorithm outperformed official Himawari-8 Level-3 hourly products across all visible bands. Compared with AERONET-OC in situ observations, root mean square errors were reduced by 34%, 26%, and 12% at 470, 510, and 640 nm, respectively. The transformer model achieved correlation coefficients above 0.98 on test data, higher than both the random forest baseline and the operational product. Comparisons with MODIS ocean color products also showed strong spatial and temporal consistency, with correlation coefficients above 0.96.

The model also corrects known retrieval biases in different water types. It reduced the underestimation of Rrs at 470 and 510 nm in turbid coastal waters while mitigating overestimation at 640 nm in clearer waters. These improvements allow the system to capture rapid optical changes within a one-hour window, changes that are often smoothed out or missed by standard hourly composites.

"Himawari-8 was not designed for ocean color observations, but our machine learning approach helps overcome its hardware limitations," said Prof. SHI Chong from AIR, corresponding author of the study. "By learning from the high-quality MODIS record, the transformer model can effectively correct sensor noise and retrieval biases. This allows us to monitor coastal water optics at 10-minute intervals, providing valuable information for fisheries, algal bloom, and water quality monitoring."

The study demonstrates that accurate 10-minute Rrs retrieval can be achieved using observations from a geostationary meteorological satellite. The algorithm effectively extends the application of Himawari-8 beyond weather observation, enabling high-frequency ocean color monitoring and providing a new tool for tracking coastal dynamics across the Asia-Pacific region.

Looking ahead, the team said they plan to expand the training dataset, improve the model's generalizability under different water and atmospheric conditions, and incorporate sun-glint correction. The framework may also be applied to other geostationary satellites, such as GK-2A and FY-4, and could support future high-frequency ocean color products by combining geostationary, polar-orbiting, and hyperspectral satellite observations.